The function of the Periaxin gene during nerve repair in a model of CMT4F

Abstract

Mutations in the Periaxin (PRX) gene are known to cause autosomal recessive demyelinating Charcot-Marie-Tooth (CMT4F) and Dejerine-Sottas disease. The pathogenesis of these diseases is not fully understood. However, progress is being made by studying both the periaxin-null mouse, a mouse model of the disease, and the protein-protein interactions of periaxin. L-periaxin is a constituent of the dystroglycan-dystrophin-related protein-2 complex linking the Schwann cell cytoskeleton to the extracellular matrix. Although periaxin-null mice myelinate normally, they develop a demyelinating peripheral neuropathy later in life. This suggests that periaxin is required for the stable maintenance of a normal myelin sheath. We carried out sciatic nerve crushes in 6-week-old periaxin-null mice, and, 6 weeks later, found that although the number of myelinated axons had returned to normal, the axon diameters remained smaller than in the contralateral uncrushed nerve. Not only do periaxin-null mice have more hyper-myelinated axons than their wild-type counterparts but they also recapitulate this hypermyelination during regeneration. Therefore, periaxin-null mice can undergo peripheral nerve remyelination, but the regulation of peripheral myelin thickness is disrupted.

Fig. 1

Myelinated axon number in periaxin-null mice is normal at baseline (uncrushed) or 2 and 6 weeks after sciatic nerve crush (crushed). Values shown are means ± SEM for 3–5 experimental mice.

Fig. 2

The cross-sectional area of periaxin-null sciatic nerve is significantly larger than that of the wild-type. The density of myelinated axons is significantly reduced in periaxin-null sciatic nerve. Values shown are means ± SEM for 19 wild-type and 23 periaxin-null mice. Asterisks indicate statistically significant differences (P > 0.001, unpaired t-test).

Fig. 3

Light micrographs of wild-type (left) and periaxin-null (right) uncrushed sciatic nerve at 12 weeks of age show the reduced density of myelinated axons and increased extracellular matrix in the periaxin-null mice. Scale bar = 10 μm.

Fig. 4

Axon diameter frequency distribution in sciatic nerve from wild-type and periaxin-null mice in uncrushed nerves, and 2 and 6 weeks after crush, respectively. The percentage frequency is plotted against axon diameter (μm). Uncrushed periaxin-null nerves have a higher percentage of small diameter fibres. After crush, the proportion of smal diameter axons increases in both wild-type and periaxin-null nerves and although this reverts slowly towards normal, at 6 week there are still more smaller axons. Values shown are means ± SEM of an average of 957 axons per each of 3–5 sciatic nerves for ea h group.

Fig. 5

Graphs of the mean g-ratio plotted against axon diameter for wild-type and periaxin-null mice at 2 and 6 weeks after crush. The axon diameter values are grouped according to size, and the mean g-ratio is plotted for each size range. For ease of comparison, the mean g-ratio is plotted against the natural log of the axon diameter, giving a straight regression line. The equations for the regression lines are displayed. Mean g-ratios are for from an average of 957 axons for each of 3–5 sciatic nerves in each experimental group. The mean g-ratio values for the crushed and uncrushed wild-type mice are superimposed at both 2 and 6 weeks after nerve crush and show that remyelination is carried out to completeness by 2 weeks after crush. The mean g-ratio values for the uncrushed periaxin-null nerves are lower than their wild-type equivalents, indicating hypermyelination. After nerve crush, the periaxin-null axon g-ratios are very similar to their wild-type counterparts at 2 weeks, indicating a similar initial rate of remyelination. However, at 6 weeks after crush, the g-ratio values have decreased, indicating progressive hypermyelination, moving closer to the hypermyelinated state of the uncrushed nerves.

Fig. 6

Graphs of the g-ratio for each axon measured in crushed and uncrushed nerve, plotted against its axon diameter, for one experimental animal for each group. This emphasizes that there is no difference in g-ratio values for crushed and uncrushed wildtype axons at 2 or 6 weeks after crush, indicating complete remyelination. However, there are distinct populations of g-ratio values from crushed and uncrushed periaxin-null axons, at 2 weeks after crush. The uncrushed values are lower, indicating hypermyelination, and the crushed values are similar to that of crushed wild-type axons at this timepoint. At 6 weeks after crush, the crushed and uncrushed periaxin-null axon g-ratio values are more similar, and lower, indicating progressive hypermyelination.